Biologically Reasoned Point-of-Interest Image Compression for Mobile Robots
In this paper authors describe image compression based on the idea of biological “yellow spot” in which the quality/resolution is variable, depending on the distance from the point-of-interest. Reducing the amount of data in a robot’s vision system enables to use a computer cluster for non-time-critical “mental” processing tasks like “memories” or “associations”. This approach can be particularly useful in HTM-based data processing of robot’s vision system data.
KeywordsMobile Robot Object Recognition Image Compression Threshold Function Computer Cluster
Unable to display preview. Download preview PDF.
- 1.Hecht, E.: Optics, 4th edn. Addison-Wesley, San Francisco (2002)Google Scholar
- 2.Hawkins, J., Dileep, G.: Hierarchical temporal memory- concepts, theory and terminology. Numenta Inc. (2007), http://www.numenta.com/Numenta_HTM_Concepts.pdf (accessed March 2, 2009)
- 3.Hawkins, J., Blakeslee, S.: On intelligence. Times Books, New York (2004)Google Scholar
- 4.Hawkins, J.: Learn like a human. IEEE Spectrum 44(7) (2007), http://www.spectrum.ie-ee.org/apr07/4982 (accessed March 2, 2009)
- 6.Podpora, M.: Biologically reasoned machine vision: RLE vs. entropy-coding compression of DWT-transformed images. In: Proc. EEICT Conference, Brno (2008)Google Scholar
- 7.Numenta Inc. Numenta Pictures Demonstration Program (2007), http://www.numenta.com/about-numenta/technology/pictures-demo.php (accessed March 2, 2009)
- 8.Sadecki, J.: Parallel optimization algorithms and investigation of their efficiency: parallel distributed memory systems. Internal Report Opole University of Technology, Opole (2001) (in Polish)Google Scholar
- 9.Baker, M.: Cluster Computing White Paper (2001), http://arxiv.org/abs/cs/0004014 (accessed March 2, 2009)
- 10.Nałęcz, M., Duch, W., Korbicz, J., Rutkowski, L., Tadeusiewicz, R.: Biocybernetics and biomedical engineering neural networks. Akademicka Oficyna Wydawnicza Exit, Warsaw 6 (2000) (in Polish)Google Scholar